Anthropic has officially announced the release of a new report detailing two potential future scenarios regarding the AI leadership competition between the United States and China in 2028. The report emphasizes that the U.S. and its allies must maintain a technological edge in the face of threats from the Chinese Communist Party (CCP) to ensure that democratic nations retain their dominant position in the development and application of AI.
Two Scenarios for the U.S. and China in 2028
The report describes two scenarios: the first scenario depicts the U.S. successfully maintaining its computational advantage, with policymakers strengthening export controls to prevent technology theft by China and accelerating the adoption of AI by democratic nations. In this scenario, democratic countries would be able to set the rules and standards for AI and potentially engage in effective dialogue with China on security issues.
“The U.S. and its allies must maintain a technological edge in the face of the threat posed by the Chinese Communist Party.”
Anthropic
The second scenario outlines a situation where the U.S. fails to take action, and policymakers do not strengthen controls over CCP's computational resources, allowing Chinese AI companies to rapidly catch up and even surpass the U.S. In this scenario, AI regulations and standards would be dominated by authoritarian regimes, with the most advanced models being used for large-scale automated repression.
Ensuring Democratic Nations' Dominance in AI Development
The report highlights that democratic nations currently possess a significant advantage in computational resources, which is crucial for developing cutting-edge AI models. This advantage stems from innovation in the U.S. and its allies, along with bipartisan consensus on export control policies. Nevertheless, the progress made by Chinese AI labs in model intelligence cannot be overlooked, making policy action increasingly urgent.
“If the most advanced AI systems are dominated by regimes that view them as tools of repression, the chances of successful transformation will be greatly diminished.”
Anthropic
The report indicates that with the rapid expansion of computational resources, the acceleration of AI capabilities will enter a new era. If the U.S. and its allies can take timely action, they may secure a technological advantage of 12 to 24 months before 2028, which will be critical for the future development of global AI safety and governance.
Addressing the Challenges of Authoritarian AI
The report concludes by emphasizing that the future development of AI will have profound implications for the economy and society, and the responsibility to ensure that this process is safe and beneficial for humanity should rest with democratically elected governments, not authoritarian regimes. If led by regimes that use AI as a tool of repression, significant threats to the transformation process will arise.
“Our mission is to ensure that humanity can safely and beneficially navigate the transition to transformative AI.”
Anthropic
Source: Anthropic Official Announcement